AlignmentResearch
/

probe-inference-weights

activation-probes

interpretability

Model card Files Files and versions

probe-inference-weights

437 MB

Ctrl+K

Ctrl+K

2 contributors

History: 5 commits

afspies-far's picture

Convert probe_metadata.json to schema 1 (input_window, readout)

0e3d386 verified about 21 hours ago

nemotron-3-nano-30b-a3b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
nemotron-3-super-120b-a12b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
qwen3.5-122b-a10b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
qwen3.5-2b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
qwen3.5-397b-a17b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
qwen3.5-9b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
qwen3.6-27b
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago
.gitattributes

1.52 kB
initial commit 7 days ago
LICENSE

1.07 kB
MIT licence for our contribution; card wording 7 days ago
LICENSE-NVIDIA-NEMOTRON-OPEN-MODEL.txt

7.95 kB
Add probe weights: 7 models x {linear, mlp, efc, axial} 7 days ago
LICENSE-QWEN-APACHE-2.0.txt

11.5 kB
Add probe weights: 7 models x {linear, mlp, efc, axial} 7 days ago
NOTICE

652 Bytes
Add probe weights: 7 models x {linear, mlp, efc, axial} 7 days ago
README.md

4.28 kB
Convert probe_metadata.json to schema 1 (input_window, readout) about 21 hours ago